experience-lwc-accessibility-jest-run
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill constructs shell commands using user-provided inputs such as module names and file paths, which creates a surface for potential command injection if the input contains shell metacharacters.
- Ingestion points: Untrusted data enters the context through user-specified values for
{moduleName},{relativePath}, and{coreRootPath}used in shell command generation withinSKILL.mdandreferences/running-sa11y-jest-tests.md. - Boundary markers: There are no explicit boundary markers or instructions to the agent to sanitize or ignore instructions embedded within the user-provided paths.
- Capability inventory: The skill generates templates for
bazel test,npm test,find,ls, andbazel querywhich interact with the system environment. - Sanitization: No sanitization or validation logic is provided for the interpolated user strings.
- [COMMAND_EXECUTION]: The skill provides templates for executing CLI tools including
bazelandnpm. These commands are standard for the described workflow but involve shell execution with parameters derived from user input, which could be misused if not properly handled by the agent.
Audit Metadata